1 Programming Languages Memory Management Chapter 11

2 Definitions Memory management: the process of binding values to memory locations. The memory accessible to a program is its address space, represented as a set of values {0, 1, …, n}. –The numbers represent memory locations. –These are logical addresses – do not always correspond to physical addresses at runtime. The exact organization of the address space depends on the operating system and the programming language being used.

3 Runtime memory management is an important part of program meaning. –The language run-time system creates & deletes stack frames, creates & deletes dynamically allocated heap objects – in cooperation with the operating system Whether done automatically (as in Java or Python), or partially by the programmer (as in C/C++), dynamic memory management is an important part of programming language design.

4 Review Definitions Method: any subprogram (function, procedure, subroutine) – depends on language terminology. Environment of an active method: the variables it can currently access plus their addresses (a set of ordered pairs) State of an active method: variable/value pairs

5 Three Categories of Memory (for Data Store) Static: storage requirements are known prior to run time; lifetime is the entire program execution Run-time stack: memory associated with active functions –Structured as stack frames (activation records) Heap: dynamically allocated storage; the least organized and most dynamic storage area

6 Static Data Memory Simplest type of memory to manage. Consists of anything that can be completely determined at compile time; e.g., global variables, constants (perhaps), code. Characteristics: –Storage requirements known prior to execution –Size of static storage area is constant throughout execution

7 Run-Time Stack The stack is a contiguous memory region that grows and shrinks as a program runs. Its purpose: to support method calls It grows (storage is allocated) when the activation record (or stack frame) is pushed on the stack at the time a method is called (activated). It shrinks when the method terminates and storage is de-allocated.

8 Run-Time Stack The stack frame has storage for local variables, parameters, and return linkage. The size and structure of a stack frame is known at compile time, but actual contents and time of allocation is unknown until runtime. How is variable lifetime affected by stack management techniques?

9 Heap Memory Heap objects are allocated/deallocated dynamically as the program runs (not associated with specific event such as function entry/exit). The kind of data found on the heap depends on the language –Strings, dynamic arrays, objects, and linked structures are typically located here. –Java and C/C++ have different policies.

10 Heap Memory Special operations (e.g., malloc, new) may be needed to allocate heap storage. When a program deallocates storage ( free, delete ) the space is returned to the heap to be re-used. Space is allocated in variable sized blocks, so deallocation may leave “holes” in the heap (fragmentation). –Compare to deallocation of stack storage

11 Heap Management Some languages (e.g. C, C++) leave heap storage deallocation to the programmer –delete Others (e.g., Java, Perl, Python, list- processing languages) employ garbage collection to reclaim unused heap space.

12 The Structure of Run-Time Memory Figure 11.1 These two areas grow towards each other as program events require.

13 Stack Overflow The following relation must hold: 0 ≤ a ≤ h ≤ n In other words, if the stack top bumps into the heap, or if the beginning of the heap is greater than the end, there are problems!

14 Heap Storage States For simplicity, we assume that memory words in the heap have one of three states: –Unused: not allocated to the program yet –Undef: allocated, but not yet assigned a value by the program –Contains some actual value

15 Heap Management Functions new returns the start address of a block of k words of unused heap storage and changes the state of the words from unused to undef. –n ≤ k, where n is the number of words of storage needed; e.g., suppose a Java class Point has data members x,y,z which are float s. –If floats require 4 bytes of storage, then Point firstCoord = new Point( ) calls for 3 X 4 bytes (at least) to be allocated and initialized to some predetermined state.

16 Heap Overflow Heap overflow occurs when a call to new occurs and the heap does not have a contiguous block of k unused words So new either fails, in the case of heap overflow, or returns a pointer to the new block

17 Heap Management Functions delete returns a block of storage to the heap The status of the returned words are returned to unused, and are available to be allocated in response to a future new call. One cause of heap overflow is a failure on the part of the program to return unused storage.

18 The New (5) Heap Allocation Function Call: Before and After Figure 11.2 A before and after view of the heap. The “after” shows the affect of an operation requesting a size-5 block. (Note difference between “undef” and “unused”.) Deallocation reverses the process.

19 Heap Allocation Heap space isn’t necessarily allocated and deallocated from one end (like the stack) because the memory is not allocated and deallocated in a predictable (first-in, first-out or last-in, first-out) order. As a result, the location of the specific memory cells depends on what is available at the time of the request.

20 Choosing a Free Block The memory manager can adopt either a first-fit or best-fit policy. Free list = a list of all the free space on the heap: 4 bytes, 32 bytes, 1024 bytes, 16 bytes, … A request for 14 bytes could be satisfied –First-fit: from the 32-byte block –Best-fit: from the 16 byte block

21 Virtual versus Physical The view of a process address space as a contiguous set of bytes consisting of static, stack, and heap storage, is a view of the logical (virtual) address space. The physical address space is managed by the operating system, and may not resemble this view at all. –OS is responsible for mapping virtual memory to physical memory and determining how much physical memory a program can have at a time. –Language is responsible for managing virtual/logical memory

22 Pointers Pointers are addresses; i.e., the value of a pointer variable is an address. Memory that is accessed through a pointer is dynamically allocated in the heap Java doesn’t have explicit pointers, but reference types are represented by their addresses and their storage is allocated on the heap (although the reference is on the stack).

: Dynamic Arrays In addition to simple variables (ints, floats, etc.) most imperative languages support structured data types. –Arrays: “[finite] ordered sequences of values that all share the same type” –Records (structs): “finite collections of values that have different types”

24 Java versus C/C++/etc. In Java, arrays are always allocated dynamically from heap memory. In many other languages –Globally defined arrays - static memory. –Local (to a function) arrays are - stack storage. –Dynamically allocated arrays - heap storage. Dynamically allocated arrays also have storage on the stack – a reference (pointer) to the heap block that holds the array.

25 Declaring Arrays Typical Java array declarations: –int[] arr = new int[5]; –float[][] arr1 = new float [10][5]; –Object[] arr2 = new Object[100]; Typical C/C++ array declarations –int arr[5]; –float arr1[10][15]; –int *intPtr; intPtr = new int[5]

26 When Heap Allocation is Needed Consider the declaration int A(n); Since array size isn’t known until runtime, storage for the array can’t be allocated in static storage or on the run-time stack. The stack contains the dope vector for the array, including a pointer to its base address, and the heap holds the array values, in contiguous locations.

27 Array Allocation and Referencing The dope vector has information needed to interpret array references: –Array base address –Array size (number of elements) for multi-dimensioned arrays, size of each dimension –Element type (which indicates the amount of storage required for each element) For dynamically allocated arrays, this information should be stored in memory to access at runtime.

28 Allocation of Stack and Heap Space for Array A Figure 11.3

29 Program Semantics for Arrays skip to slide 39 Semantics = program meaning If State is the set of all program states, the meaning M of an abstract Clite Program is defined by –M: Program → State –M: Statement X State → State –M: Expression X State → Value

30 Program Semantics M: Program → State –The meaning of a program is a function that produces a state. M: Statement X State → State –The meaning of a statement is a function that, given a current state, yields a new state M: Expression X State → Value –The meaning of an expression is a function that, given a current state, yields a value.

31 Example For the Clite abstract syntax rule Program = Declarations decpart; Block body the meaning rule is as follows: “The meaning of a Program is defined to be the meaning of its body when given an initial state consisting of the variables of the decpart, each initialized to the undef value corresponding to its declared type.” (page 200)

32 1-D Array Semantics Notation From Chapter 5, page 116: –addr(a[i]) = addr(a[0]) + e∙i where e is element size From Chapter 2, page 53, abstract syntax –ArrayRef = String id ; Expression index ; –ArrayDecl = Variable v ; Type t ; Integer size ; –For ArrayDecl ad, ad.size = # of elements –For ArrayRef ar, ar.index = value of index expression

33 Array Semantics Assume: –Array is dynamically declared –Array is one dimension only –Array element size is 1 (word) –Array indexing is 0-based ad is the array declaration, ar is an array reference.

34 Meaning Rule 11.1: The meaning of an ArrayDecl ad is: 1. Compute addr(ad[0]) = new(ad.size). (Allocate enough storage to hold size elements of ad.type. addr(ad[0])= start address of the block of storage) 2. Push addr(ad[0 ] ) onto the stack. 3. Push ad.size onto the stack. 4. Push ad.type onto the stack. Step 1 creates a heap block for ad. Steps 2-4 create the dope vector for ad in the stack.

35 Implementing the Meaning Rule The compiler generates code to perform the steps outlined in the meaning rule and incorporates them into the object code wherever there is an array declaration. If ‘ new ’ fails, an exception is generated.

36 Meaning Rule 11.2 The meaning of an ArrayRef ar for an array declaration ad is: (assume element size is 1) 1.Compute addr( ad[ar.index] ) = addr( ad[0] ) + ( ad.index - 1] ) (where ad.index-1 = ad[index – 1]) 2. If addr( ad[0] )  addr( ad[ar.index] ) < addr (ad[0])+ad.size, then return the value at addr( ad[ar.index] ) 3. Otherwise, signal an index-out-of-range error.

37 Array Assignments: a[i] = Expr Meaning Rule 11.3 The meaning of an array Assignment as is: 1.Compute addr( ad[ar.index] )=addr( ad[0] ) +( ad.index-1 ) 2.If addr( ad[0] )  addr( ad[ar.index] ) < addr (ad[0])+ad.size) then assign the value of as.source to addr( ad[ar.index] ) (the target) 3.Otherwise, signal an index-out-of-range error.

38 Example The assignment A[5]=3 changes the value at heap address addr(A[0])+4 to 3, since ar.index=5 and addr(A[5])=addr(A[0])+4. This assumes that the size of an int is one word.

39 Alternative Storage Allocation for Arrays and Structs C/C++ support static (globally defined) arrays C/C++ also have fixed stack-dynamic arrays –Arrays declared in functions are allocated storage on the stack, just like other local variables. –Index range and element type are static Ada also permits (variable) stack-dynamic arrays –Index range can be specified as a variable Get(List_Len); Declare List: array (1.. List_Len) of Integer;

Memory Leaks and Garbage Collection The increasing popularity of OO programming has meant more emphasis on heap storage management. Active objects: can be accessed through a pointer or reference. Inactive objects: blocks that cannot be accessed; no reference exists. (Accessible and inaccessible may be more descriptive.)

41 Review Three types of storage –Static –Stack –Heap Problems with heap storage: –Memory leaks (garbage): failure to free storage when pointers (references) are reassigned –Dangling pointers: when storage is freed, but references to the storage still exist.

42 Allocation of Stack and Heap Space for Array A Figure 11.3

43 Garbage Garbage: any block of heap memory that cannot be accessed by the program; i.e., there is no stack pointer to the block; but which the runtime system thinks is in use. Garbage is created in several ways: –A function ends without returning the space allocated to a local array or other dynamic variable. The pointer (dope vector) is gone. –A node is deleted from a linked data structure, but isn’t freed –…

44 Another Problem A second type of problem can occur when a program assigns more than one pointer to a block of heap memory The block may be deleted and one of the pointers set to null, but the other pointers still exist. If the runtime system reassigns the memory to another object, the original pointers pose a danger.

45 Terminology A dangling pointer (or dangling reference, or widow) is a pointer (reference) that still contains the address of heap space that has been deallocated (returned to the free list). An orphan (garbage) is a block of allocated heap memory that is no longer accessible through any pointer. A memory leak is a gradual loss of available memory due to the creation of garbage.

46 Widows and Orphans Consider this code: class node { int value; node next; }... node p, q; p = new node(); q = new node();... q = p; delete(p); The statement q = p; creates a memory leak. The node originally pointed to by q is no longer accessible – it’s an orphan (garbage). Now, add the statement delete(p); The pointer p is correctly set to null, but q is now a dangling pointer (or widow)

47 Creating Widows and Orphans: A Simple Example Figure 11.4 (a): after new(p); new(q); (b): after q = p; (c): after delete(p); q still points to a location in the heap, which could be allocated to another request in the future. The node originally pointed to by q is now garbage.

48 Python Memory Allocation A 3.5 A4 7.0 A = A * 2 A = “cat” cat Python may allocate new storage with each assignment, so it handles memory management automatically. It will create new objects and store them in memory; it will also execute garbage collection algorithms to reclaim any inaccessible memory locations. Variables contain references to data values

Garbage Collection All inaccessible blocks of storage are identified and returned to the free list. The heap may also be compacted at this time: allocated space is compressed into one end of the heap, leaving all free space in a large block at the other end.

50 Garbage Collection C & C++ leave it to the programmer – if an unused block of storage isn’t explicitly freed by the program, it becomes garbage. –You can get C++ garbage collectors, but they aren’t standard Java, Python, Perl, (and other scripting languages) are examples of languages with garbage collection –Python, etc. also automatic allocation: no need for “new” statements Garbage collection was pioneered by languages like Lisp, which constantly creates and destroys linked lists.

51 Implementing Automated Garbage Collection If programmers were perfect, garbage collection wouldn’t be needed. However,... There are three major approaches to automating the process: –Reference counting –Mark-sweep –Copy collection

52 Reference Counting Initially, the heap is structured as a linked list (free list) of nodes. Each node has a reference count field; initially 0. When a block is allocated it’s removed from the free list and its reference count is set to 1. When pointers are assigned or freed the count is incremented or decremented. When a block’s count goes back to zero, return it to the free list and reduce the reference count of any node it points to.

53 P Q 21 “P = null” reduces the reference count of the first node to 1 “Q = null” reduces the reference count of the first node to 0, which triggers the reduction of the reference count in node 2 to 0, recursively reduces the ref. count in node 3 to 0, and then returns all three nodes to the free list. A Simple Illustration 1null

54 Node Structure and Example Heap for Reference Counting Figure 11.5 There’s a block at the bottom whose reference count is 0. What does this represent? What would happen if delete is performed on p and q?

55 Reference Counting: Pros and Cons Advantage: the algorithm is performed whenever there is an assignment or other heap action. Overhead is distributed over program lifetime Disadvantages are: –Can’t detect inaccessible circular lists. –Extra overhead due to reference counts (storage and time).

56 Mark and Sweep Runs when the heap is full (free list is empty or cannot satisfy a request). Two-pass process: –Pass 1: All active references on the stack are followed and the blocks they point to are marked (using a special mark bit set to 1). –Pass 2: The entire heap is swept, looking for unmarked blocks, which are then returned to the free list. At the same time, the mark bits are turned off (set to 0).

57 Mark Algorithm Mark(R): //R is a stack reference If (R.MB == 0) R.MB = 1; If (R.next != null) Mark(R.next); All reachable nodes are marked. Starts in the stack, moves to the heap.

58 Sweep Algorithm Sweep( ): i = h; // h = first heap address While (i<=n) { if(i.MB == 0) free(i);//add node i to free list else i.MB = 0; i++; } Operates only on the heap.

59 Node Structure and Example for Mark-Sweep Algorithm Figure 11.6 Before the mark-sweep algorithm begins

60 Heap after Pass I of Mark-Sweep Figure 5.16 After the first (mark) pass, accessible nodes are marked, others aren’t

61 Heap after Pass II of Mark-Sweep Figure 11.8 After the 2 nd (sweep) pass: All inaccessible nodes are linked into a free list; all accessible nodes have their mark bits returned to 0

62 Mark and Sweep: Pros and Cons Advantages: –It may never run (it only runs when the heap is full). –It finds and frees all unused memory blocks. Disadvantage: It is very intensive when it does run. Long, unpredictable delays are unacceptable for some applications.

63 Copy Collection Similar to mark and sweep in that it runs when the heap is full. Faster than mark and sweep because it only makes one pass through the heap. No extra reference count or mark bit needed. The heap is divided into two halves, from_space and to_space.

64 Copy Collection (Stop and Copy) While garbage collection isn’t needed, –From_space contains allocated nodes and nodes on the free list. –To_space is unusable. When there are no more un-allocated nodes in from_space, “flip” the two spaces, and pack all accessible nodes in the old from_space into the new from_space. Any left-over space is the free space.

65 Initial Heap Organization for Copy Collection Figure 11.9 Not available

66 Result of a Copy Collection Activation Figure After “flipping” and repacking into the former to_space. (The accessible nodes are packed, orphans are returned to the free_list, and the two halves reverse roles.)

67 Discussion When an active object is copied to the to_space, update any references contained in the objects When copying is completed, the new to_space contains only active objects, and they are tightly packed into the space. Consequently, the heap is automatically compacted (defragmented).

68 Analysis Automatic compaction is the main advantage of this method when compared to mark-and-sweep. Disadvantages: –All active objects must be copied: may take a lot of time (not necessarily as much as the two-pass algorithm). –Requires twice as much space for the heap

69 Copy Collection v Mark-Sweep If r, the ratio of active heap blocks to heap size, is significantly less than (heap size)/2, copy collection is more efficient –Efficiency = amount of memory reclaimed per unit of time As r approaches (heap size)/2 mark-sweep becomes more efficient Based on a study reported in a paper Jones and Lins, 1996.

70 Garbage Collection Analysis Different languages and implementations will probably use some variation or combination of one of the above strategies. Java runs garbage collection as a background process when demand on the system is low, hoping that the heap will never be full. Java also allows programmers to explicitly request garbage collection, without waiting for the system to do it automatically. Functional languages (Lisp, Scheme, …) also have built-in garbage collectors C/C++ do not.

71 Garbage Collection Summarized Some commercial applications divide nodes into categories according to how long they’ve been in memory –The assumption is that long-resident nodes are likely to be permanent – don’t examine them –New nodes are less likely to be permanent – consider them first –There may be several “aging” levels